Research map / 3D vision / Gaussian splatting
From Blobs to Spokes: High-Fidelity Surface Reconstruction via Oriented Gaussians
Diego Gomez, Antoine Guedon, Nissim Maruani, Bingchen Gong, Maks Ovsjanikov
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add normal alignment loss (ablation) | Gaussian Wrapping · T&T Virtual Scan F1 | F1: 0.53 -> 0.52 | No clear effect |
| add normal alignment loss (ablation) | Gaussian Wrapping · T&T MBR | LPIPS: 0.3718 -> 0.3683 | Improved |
| add normal-aware densification on top of normal alignment loss (ablation) | Gaussian Wrapping · T&T MBR | PSNR: 20.78 -> 20.98 | Improved |
| plug Gaussian Wrapping into RaDe-GS (ablation) | RaDe-GS · T&T Virtual Scan F1 | F1: 0.39 -> 0.48 | Improved |
| use 2 pivots per Gaussian instead of 9 prior-work pivots | Gaussian Wrapping · T&T mesh size | #Verts (M): 11.79 vs 43.46 (GGGS, Mip-NeRF 360) | Improved |
| Primal Adaptive Meshing instead of pivot-based meshing | Gaussian Wrapping · T&T Uniform Sampling F1 | F1: 0.48 -> 0.43 | Hurt |
| evaluation protocol: Virtual Scanning vs Uniform Sampling for PGSR | PGSR · T&T F1 | F1: 0.52 vs 0.53 | Mixed |
| oriented Gaussians vs unoriented attenuation (GGGS) | Gaussian Wrapping · T&T Virtual Scan F1 | F1: 0.53 vs 0.53 | No clear effect |
Closest papers by meaning
- SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering (CVPR 2024 · similarity 0.91)
- VA-GS: Enhancing the Geometric Representation of Gaussian Splatting via View Alignment (NeurIPS 2025 · similarity 0.90)
- FatesGS: Fast and Accurate Sparse-View Surface Reconstruction Using Gaussian Splatting with Depth-Feature Consistency (AAAI 2025 · similarity 0.90)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.90)
- AtlasGS: Atlanta-world Guided Surface Reconstruction with Implicit Structured Gaussians (NeurIPS 2025 · similarity 0.90)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
